This project proposes to undertake a detailed analysis on data from a large epidemiologically based, multi-informant psychological autopsy study (MH 38198) of 122 adolescent suicide completers, aged under twenty, who died in the Greater Metropolitan New York- area between June, 1984 and May, 1986 and comparable numbers of age, sex and ethnically matched adolescents who attempted suicide, and community control adolescents (who have been similarly matched). The psychological autopsy comprised an extensive semi-structured interview with the adolescents and their parents and, in the case of the index group, with a similar aged peer or sibling. The domains of inquiry were psychiatric symptoms prior to the death or attempt, stressful events, exposure to suicide related events, as well as extensive information on family background and intra-familial relations. The data collected in this study are unique in that they are (1) controlled (which will enable us to determine which con-elates are specific for suicide. (2) based on a very large sample (which will enable us to examine sub-group differences in a way that has not been possible in previous, much smaller, studies), and (3) derived from a very comprehensive investigative battery (which will allow us to examine risk factors, a full range of diagnostic correlates, and facilitating or triggering factors). We expect that the proposed data analysis will provide information on the prevalence of psychiatric disorder, the importance of different types of disturbance (which can be incorporated into early warning signs for use in prevention programs), information on the distinctions between suicide completers and attempters (which should assist in identifying attempters with a high potential for later suicide death), information on stressful events which may constitute a specific hazard for suicide, information on family correlates, and information on the importance of social factors, such as the availability of firearms and exposure to provocative media material.